Maneet Mehta

I study Statistics & Data Science and Physics & Philosophy at Yale, with my attention on when institutions should hand decisions to AI and whether models can explain their own reasoning faithfully. This fall, I'm a research fellow with the Supervised Program for Alignment Research, running controlled experiments on LLM explanation faithfulness.

Recently, I've on AI task allocation in public administration at the Hertie School in Berlin, on emotional bias in image-generation models at the University of Maryland, and on NLP pipelines for defense-policy analysis with the German Council on Foreign Relations.

In my free time, I enjoy competitive debate, backpacking, cooking, pickleball, and reading Eastern philosophy.